TECHNICAL FEATURE
This article was published in ASHRAE Journal, April 2015. Copyright 2015 ASHRAE. Reprinted here by permission from ASHRAE at
www.calmac.com. This article may not be copied nor distributed in either paper or digital form by other parties without ASHRAE’s permission.
For more information about ASHRAE, visit www.ashrae.org.
The Hidden Daytime Price
Of Electricity
BY EVAN BERGER
Whether or not you know it, if you manage an office building, school, university, mall,
or hospital and are in a region that has a demand charge over $10 per kilowatt each
month, the price you pay for electricity is likely more than twice as much during the
day than it is at night. Even customers who receive “flat rates” from their utility or
third-party supplier pay a much higher rate during daytime hours, due to the effects
of demand charges. In a sense, demand charges serve as a peak-time adder for a typical nonresidential customer with a bell-shaped load curve, making energy twice as
expensive during the day – or, looking at it from another perspective, half-off at night.
Yet despite the outsized effect that demand charges have
on commercial, industrial, and institutional customers,
these costs are poorly understood by the entities who
pay them. This article provides an overview of various
“demand” charges, such as utility demand, grid demand,
and rate structure-related costs. Further, this article demonstrates how reducing peak demand and managing one’s
load curve can provide customers with opportunities to
cut costs dramatically, and also potentially benefit from
revenue-generating programs such as demand response.
Utility Demand Charges
The utility is, broadly speaking, the company that
delivers power to your home and business. It is the
“poles and wires company” that provides the last-mile
distribution to the end-user; in regulated states, it may
also be the company that owns the generators who make
your power as well.
The most commonly understood demand charges
are those levied by the utility. Many, but not all, utilities use demand charges to earn revenue; in deregulated regions, which include 16 states plus the District
of Columbia in the United States, demand charges
are a principal means by which investor-owned utilities profit from commercial and industrial customers.
Utility demand charges are typically denoted in dollars per kilowatt ($/kW) monthly, and set through a
ratemaking process between the utility and its public
Evan Berger is director of energy solutions for CALMAC Manufacturing Corp., in Fair Lawn, N.J.
64
A S H R A E J O U R N A L ashrae.org APRI L 2015
TECHNICAL FEATURE utility commission. Typically, the amount of kW assessed
on each monthly bill is based on the highest 15- or
30-minute kW interval within that billing period. Utility
demand charges are the most widely understood among
commercial customers because they are easiest to intellectually comprehend, and also very often the most
transparent form of demand charge on a customer bill.
There are several complexities to navigate when
deciphering utility demand charges, however. The
first is time-of-use or seasonal adders. Many summerpeaking utilities have summer adders: in New Jersey,
the utility PSE&G levies an annual demand charge
for commercial customers year-round; in the months
June through September there is an adder that effectively triples the total utility demand charge. Another
demand charge adder is based on time-of-use. For
example, Southern California Edison (SCE) has one
rate with a summertime on-peak demand charge
adder during the weekday hours of 12 to 6 p.m. of
$26.01/kW, as well as a mid-peak adder of $7.17/kW, for
the site’s peak during weekday hours of 8 a.m. to 12
p.m. and 6 to 11 p.m. These two charges are in addition
to SCE’s year-round off-peak demand charge of $14.32/
kW.1 All of the California investor-owned utilities have
rates that model this format.
Another source of complexity in utility demand
charges is the use of ratchets. Many utilities use ratchets to assess a higher kW number to commercial customers. Oncor, one of Texas’s largest utilities, imposes
an 80% ratchet for large users. In Oncor’s case, the
ratchet works as follows: a customer’s assessed kW is
the greater of a) its monthly peak kW draw, or b) 80%
of the peak kW draw over the course of the previous
11 billing months.2 This penalizes customers who hit
a particularly high kW peak in any given month, as
that peak can affect their electricity expenditure for
the next year. For example, if a customer has a typical
monthly peak kW of 2,500 kW, but happens to hit a peak
of 5,000 kW in the month of August, its energy costs
will be affected for the following year: even if the customer never exceeds 2,500 kW in any 15-minute interval
again, its assessed demand for the next 11 months will be
a minimum of 80% of its August peak, or 4,000 kW.
A final, and perhaps most confusing, source of complexity in utility demand charges is pricing based on
load factor. Load factor is defined as the average kW
draw divided by the peak interval.
FIGURE 1 Nine ISO/RTOs in North America. Source: FERC.
138.8 kW Average
500 kW Maximum = 27.7 Load Factor
Other utilities use a more tortuous method referred to as
hours use of demand (HUD). A simple glance at an HUDbased bill might give most customers no idea that they are
sensitive to peak demand fluctuations; but HUD actually
serves as a very expensive form of demand charge.
For its large C&I customers, Georgia Power’s PLL-9 rate
has no $/kW demand charge, but rather has an hours use
of demand structure that, if understood, functions as an
incentive for buildings to shave their daytime peak.3 The
best way to explain Georgia Power’s rate is by example.
Suppose you had a very small building – perhaps a large
doghouse or a toolshed – which has a steady 1 kW load all
month. At the end a 30-day month you would have used
720 kWh. With Georgia Power’s PLL-9 rate and a 1 kW peak
load, the first 200 “Hours Use of Demand” would charge
you 12.7 cents for the first 200 kWh and the remaining
520 kWh would be at a little more than a penny per kWh.4
Using the same example – with the exception being that
for 1 hour of the month the load went up to 2 kW and for
one hour the load was zero – the total kWh would be the
same 720, however now the first 400 kWh (2 kW x 200
HUD) would be at 12.7 cents, and the balance at 1.3 cents.
Therefore, the one single hour of increased demand
doubles the amount of electricity charged at the higher
rate. After surcharges and taxes, Georgia Power’s effective
demand charge for PLL-9 customers is in excess of $20/
kW monthly – and it is 95% ratcheted.
Grid Demand Charges
Independent System Operators (ISOs) and Regional
Transmission Operators (RTOs) are both terms to
APRI L 2015 ashrae.org A S H R A E J O U R N A L
65
$/kW
25
20
15
10
5
0
Jan FebMar Apr MayJun Jul AugSepOct Nov Dec
2014
PSE&G Annual Demand
PJM Capacity Demand
PSE&G Summer Demand
PJM Network Transmission Demand
FIGURE 2 Cost of cooling per kW in New Jersey’s PSE&G utility. Sources: PJM, PSE&G; General Power & Lighting Rate. Note:
C&I customers continue to pay for cooling during winter due to PJM’s grid demand charges, which are 100% ratcheted.
Advertisement formerly in this space.
66
describe the coordinators of the
bulk grid, the non-profit entities
that manage the dispatch of wholesale electricity in their respective
footprints.
ISO/RTOs’ role in demand
charges are poorly understood, in
large part because few consumers know of their existence, and
because their demand charges are
often collected indirectly, through
a retail electricity provider, and
thus often buried in C&I bills.
There are nine ISO/RTOs across
North America, including ISOs
in the three largest U.S. states,
California, Texas, and New York,
and regional grid operators across
the country as seen in Figure 2. The
largest bulk grid in terms of MW
managed is PJM Interconnection,
which covers the mid-Atlantic, DC
Metro area, Virginia, Ohio, and
Chicago.
One of the most important functions of the ISOs/RTOs is to ensure
that its territory has adequate generation to meet its worst-case contingency – the grid’s equivalent to a
design day. Each ISO or RTO meets
this function differently, but many do
so by procuring so-called “capacity”
A S H R A E J O U R N A L ashrae.org APRI L 2015
through an auction process for generators. Once the generators’ proceeds
are determined, the costs for capacity
are levied on consumers.
Most ISO/RTOs charge end-users
for their share of capacity costs
through a peak load contribution
(PLC), also referred to as a customer’s ICAP, capacity tag, or “captag”
for short. A customer’s PLC is not
based on its own peak kW draw;
rather, it is based on the customer’s
kW consumption during the grid’s
peak kW draw. In PJM, for example,
each customer’s PLC is based on the
average of the customer’s load during the grid operator’s five highest
hours of grid demand during the
year. These hours are known as the
five coincident peaks (5CPs), and it
should be noted that no more than
one of the 5CPs can be assessed
on any single day. In contrast,
NYISO determines each customer’s
captag based on the customer’s
consumption during the New York
State grid’s single highest hour of
demand each year. In both PJM and
NYISO, the peak hours of demand
fall almost exclusively on the hottest weekday afternoons of the
summer.
Advertisement formerly in this space.
TECHNICAL FEATURE
12 p.m.
11 p.m.
10 p.m.
9 p.m.
8 p.m.
7 p.m.
6 p.m.
5 p.m.
4 p.m.
3 p.m.
2 p.m.
1 p.m.
12 p.m.
11 a.m.
10 a.m.
9 a.m.
8 a.m.
7 a.m.
6 a.m.
5 a.m.
4 a.m.
3 a.m.
2 a.m.
1 a.m.
$0.120
ISO/RTOs’ operation and maintenance
$0.100
costs are determined through a regulatory
procedure, not an auction, but they are typi$0.080
cally also charged to the customer through a
$0.060
PLC-type process. This is the case in ERCOT,
$0.040
which uses a “4CP” calculation to determine
$0.020
its transmission cost recovery charges. In
$0.000
ERCOT, the customer’s assessed kW is based
on the average of its consumption during
the highest demand hour in each of the four
FIGURE 3 PEPCO Washington D.C. prices, average July Weekday. Source: PJM Interconnection. Average weekday Locational Marginal Price of electricity in PEPCO (Washington, D.C. area) in July 2013. Prices reached their
summer months, from June to September.
highest at 2 p.m., at 9.7 cents per kWh; the highest hour during the month was 2 p.m. on July 17th, when prices
Two additional notes are warranted on grid
reached 40.1 cents/kWh. Prices were at their lowest at 4 a.m., when they averaged 2.1 cents/kWh.
demand charges. First, they are frequently
ratcheted: the peak load contribution
that a customer sets in the summer will affect its grid
thermal storage is a cost-effective plan. As the graphic
demand charges for a full year. Second, there is often a
of summertime real-time locational marginal prices in
lag between when a customer’s peak load contribution
Washington, D.C.-area PEPCO shows, nighttime elecis assessed, and when it is levied. In PJM, NYISO, and
tricity on the real-time market tends to be much less
ERCOT, the grid demand charges that a customer must
expensive and less volatile.
pay are based on its PLC from the year before. Therefore,
It is important to note that regardless of whether a
in these regions, customers who set high PLCs in the
customer is on real-time pricing or a fixed rate, it is still
summer of 2014 will begin paying higher grid demand
likely to pay demand charges.
charges beginning in the summer of 2015.
Special Rate Structures
Since grid demand charges are frequently obscured in
Some utilities offer special rate structures for customthe customer bill, it is typically a subject worth inquiring
about with one’s utility or third-party provider, or with a bill- ers who have particularly high load factors. For example,
ing expert. These charges can be substantial, and are very Potomac Edison in West Virginia offers a special High
often blended into other charges within the electric bill. Load Factor schedule for large customers; these customCustomers that would like to identify their grid demand ers pay a slightly higher demand charge than normal
charges and find solutions to reduce those costs can often ask users, but they are compensated with a lower cents per
and receive for the ISO/RTO demand charges to be separated kWh usage charge.5 This benefits industrial customers
with 24x7 operations and a flat load curve: the increase
and listed on their bill in an itemized, line-by-line format.
in demand charges is offset many times over by the
Real-Time Pricing: Following the Market
decrease in cents/kWh.
Real-time pricing, also known as “indexed” or “floatOther special rate structures are given to customers
ing” rates, allow end-users to buy energy at the prewith specific technologies, such as thermal energy storvailing market price, rather than at a fixed cents per
age (TES). Austin Energy in Texas offers a special rate to
kilowatt-hour price. Over the long run, this tends to be
end-users with thermal storage onsite.6 In exchange for
a slightly higher demand charge, customers in the speless expensive: fixed pricing is an insurance mechacial TES Rate pay lower cents/kWh for most of the year,
nism, and insurance typically comes at a premium.
However, employing real-time pricing leaves customers particularly during nighttime hours; in the summer
months (June through September), TES rate customers
more exposed to the vagaries of the open market: if the
weather is unexpectedly hot or cold, or a large generator pay only 2.7 cents/kWh between 10 p.m. and 6 a.m., in
contrast to the typical user’s 6.4 cents/kWh. Since TES
fails, prices can spike dramatically.
users consume a significant amount of electricity at
For customers willing to manage the risk of cheaper
night, when they are charging up their thermal storage
but spikier real-time pricing, shifting load from day
systems, the TES rate provides immense savings.
to night with smart building technologies such as
68
A S H R A E J O U R N A L ashrae.org APRI L 2015
Advertisement formerly in this space.
TECHNICAL FEATURE
Demand Response: Turning the Paradigm Upside Down
Most of this article focuses on how demand costs consumers money; with demand response, consumers can
use their load flexibility to earn money from revenuegenerating programs run at the ISO/RTO or utility level.
An example of this is PJM’s capacity market: users with
flexible load can bid demand reduction (known as
demand response) into the capacity market and receive
the same value for their flexibility as a generator does. In
return for the revenue they receive, demand response
participants must respond to grid or utility “events” – calls
to curtail – by reducing their electricity load, typically for
a period lasting four or six hours. Such programs can be
found throughout the country, and can be very lucrative:
in New York City, combining ConEdison’s and NYISO’s
Demand Response programs can earn a customer as
much as $250,000 per curtailable megawatt, annually.
There are other demand response-related programs
available to end-users aside from the traditional curtailment programs. One such program garnering attention
is behind-the-meter frequency regulation, whereby
assets follow an ongoing grid signal to maintain the
grid at its desired frequency of 60 Hertz. Traditionally,
only natural gas and hydroelectric generators were dispatched to regulate grid frequency; but now, smaller
customer-sided assets such as electrical and thermal
storage as well as variable frequency drives (VFDs) are
engaging in this lucrative program as well. Part of this
change has been spurred by policy: FERC’s Order 755 in
2011 mandated that faster-acting devices, such as storage
and VFDs, should be paid an additional “performance”
payment for responding to grid signals more quickly
than large generators can.
Conclusion
Whether their energy managers know it or not, most C&I
buildings pay a large percentage of their electricity costs
in the form of demand charges. Nonresidential customers
with bell-shaped load curves, including office buildings,
hospitals, schools, universities, and many industrial plants,
pay substantially more during the daytime than they do at
night, because of the effect of demand charges.
Advertisement formerly in this space.
70
A S H R A E J O U R N A L ashrae.org APRI L 2015
Advertisement formerly in this space.
TECHNICAL FEATURE
If you could fill your car up with gasoline at noon for
$2.50 per gallon, or at 9 p.m. for $1.25 per gallon, which
would you choose? Many electricity customers face this
same choice – and because of the complexities of the
electricity bill, they have no idea they ever had a choice
in the first place.
Smart building technologies allow commercial, industrial, and institutional customers to buy electricity at
night, when it is vastly cheaper. Thermal energy storage,
such as ice storage and chilled water storage, allow endusers to build up cooling at night, when electricity is “on
sale,” and dispatch that cooling during the daytime to
displace the on-peak usage of their chillers and HVAC
equipment. Other demand-limiting technology, such
as battery storage or smart building controls, offer the
same opportunity to shift loads from daytime to nighttime, when electricity can be procured at a discount.
Because of the great but latent sensitivity of end-users to
demand, such load-shifting can result in electricity savings of 10% to 20% off of the total bill.
Advertisement formerly in this space.
To save money, end-users and their energy advisors
must be able to understand the effects of demand charges
– and yet, many energy procurers do not understand
that, unlike a residential bill, most C&I bills have demand
charges of one form or another. While this knowledge
gap cannot be bridged with a single article, perhaps a few
final points will be useful to professionals looking to lower
their demand charges or those of their customers.
First, all else being equal, reducing daytime peaks lowers demand costs, and thus lowers electricity bills. Smart
building technologies such as thermal energy storage help
to reduce peaks and flatten the load curve; they should be
considered in all new construction and retrofit designs.
Second, it always helps to look closely at the electricity bill. Specifically, it is of great value to add up the
demand charges (the per-kW line items) and compare
them to the usage charges (the per-kWh line items). This
gives one a sense of a site’s sensitivity to demand. When
looking at electricity bills, be sure to review at least one
bill from the summer (July or August) and one from the
winter (December, January, or February). Comparing
these two will determine whether there are seasonal
adders, and those can be a very substantial portion of
a user’s bill. However, make sure to be wary of thirdparty suppliers’ “flat rate” or “blended rate”: these terms
can obscure the effect of demand, and give the wrong
impression that electricity is equally expensive during
day and night. Rarely is this true.
Finally, when in doubt, ask for help. Non-residential
customers have access to utility representatives who can
guide them through the process of understanding their
bills; additionally, third-party electricity suppliers can
be a valuable resource as well.
References and Notes
1. Southern California Edison. 2013. Schedule TOU-8, Time-OfUse – General Service – Large.
2.Oncor Electric Delivery Company. 2014. Tariff for Retail Delivery Service, Secondary Service Greater than 10 kW.
3.Georgia Power. 2014. Electric Service Tariff, Power and Light
Large Schedule: “PLL-8.”
4.This does not include the Fuel Cost Recovery surcharge (FCR23), nor other Riders such as Environmental Compliance and
Nuclear Construction Cost Recovery. Also note that PLL-9 covers
loads 500 kW or greater; this highly simplified 1 kW site example is
for expository purposes.
5.The Potomac Edison Company. 2014. Rates and Rules & Regulations for Electric Service in Certain Counties in West Virginia,
Light and Power Service (High Load Factor) Schedule “PH.”
6.Austin Energy. 2014. City of Austin Electric Rate Schedules,
Thermal Energy Storage.
72
A S H R A E J O U R N A L ashrae.org APRI L 2015